[userbinator]

So I have no idea what it means to have everybody 'pass'.

That reminds me of this:

https://en.wikipedia.org/wiki/Grading_systems_by_country

It's interesting to see where the 'pass' bar is in different countries. I've always found the 50% mark, which seems quite popular, to be rather unusual since it implies that someone who 'passed' essentially was correct on only half the material tested (which is then a fraction of the material actually taught), and in any other setting a 50% failure rate would be completely unacceptable.

[EliRivers]

Running a little off-topic, there is evidence that a Cambridge maths exam series 150 years ago had a pass mark of about one and a half percent, and the top scorer achieved about 45 percent.

https://en.wikipedia.org/wiki/Mathematical_Tripos

[cafard]

According to Ulam's memoirs, Hardy thought very poorly of the Tripos as a test of anything in particular.

[jakobegger]

Most real life situations don't require you to solve 3 distinct problems in an hour.

That's why most degrees require you to produce a thesis, which is closer to what you'll need to do later.

[Rylinks]

I have had many tests where the material tested is not a strict subset of the material taught, and some problems require creative generalization or insight during the test.

[pandler]

I had this too in Mechanical Engineering, especially on exams where you have e.g. an hour and a half to solve 5 problems and more so as you take higher level classes. There would often be room for interpretation in the questions, and part of the problem was to make sure you're making the right assumptions about the system.

I remember one test in particular in a class about heat transfer where one question was essentially "Same problem as the previous question, but now assume <something less simplistic>". My answer was something along the lines of, "Whoops, I already made that assumption in the previous question. Extra points?" The professor was kind enough to grant me a few bonus points.

[dbcurtis]

Life 101?